Measuring Call Attempt Frequency Against a Written Rule

Research question and evidence scope

Research question: how many attempts does a campaign make to one person within a defined window, and does the count match the written frequency rule? The unit is one person or number, not one call. Attempts across campaigns, channels, and systems may need to be combined before the rule can be tested.

Start with the written rule. It should state the maximum attempts, the window, the reset event, the treatment of unanswered calls, and the exceptions. If the rule says three attempts in seven days, define whether a voicemail counts, whether a wrong number resets the clock, and whether a customer request changes the count.

Build the person-level timeline from the dialer and CRM. Link records that share a stable identifier, not only a phone number. A person may have two numbers, and a number may be shared. Record the linking evidence and label uncertain links. Do not merge records just to produce a cleaner count.

Operating scenario and evidence

Capture the attempt time, the campaign, the list source, the outcome, the agent, the disposition, and the next scheduled attempt. Include manual calls, callbacks, and attempts from another system if the rule covers them. An incomplete timeline will understate the frequency and overstate compliance.

Define the counting rule precisely. Does an attempt count when the dialer releases the call, when it connects, or when an agent speaks? Does a customer-requested callback count toward the limit? Does a failed technical attempt count? Write the answers and apply them consistently.

Method and review criteria

Sample people who received the most attempts, a random set of routine records, and a set with mixed outcomes. Include records with multiple numbers and records with a recent opt-out. Compare the observed count with the rule and record every exception with its owner.

Run two reviewers on the same timelines. Compare the linked records, the counted attempts, and the reset decision. A disagreement about whether two numbers belong to one person is a data quality finding, not a reason to guess. Route the ambiguity to the list owner.

Test sensitivity to the linking rule, the channel scope, and the reset event. If the compliance conclusion changes when a manual call is included or a second number is linked, report that dependency. The buyer needs to know which decision relies on the narrower reading.

Authority and practical implications

Report the number of people reviewed, the attempts observed, the rule limit, the exceptions, and the unresolved links. Pair the rate with complaints, opt-outs, and answer rates. A campaign can be within the numeric limit and still create an unacceptable customer experience.

The method does not determine the legal limit or certify compliance. It tests whether the operation follows its own written rule and whether the data can support the claim. Qualified counsel and the business owner decide the applicable rule.

Record linking is the hardest part. A shared household number, a reassigned business line, or a person with two mobiles can change the count. Use the strongest available identifier and label uncertain links. Do not merge records just to make the timeline look complete. If the link is uncertain, analyze the cases separately and report the ambiguity.

Limitations, conclusion, and sources

The decision record should state the rule version, the linking method, the channels included, the window, the reset event, the exceptions, and the owner. It should show the people reviewed and the attempts observed. If the campaign is within the limit but the customer experience is poor, the record should say that too.

Exception review should be separate from the main count. A customer-requested callback, a wrong number, or a technical failure may justify an extra attempt. Record the reason and the owner. If exceptions become common, the rule may be unrealistic or the data may be wrong.

The operational use is to repair the list, the dialer schedule, or the rule. A campaign that repeatedly exceeds the limit needs a pause, not a reminder to the callers. The decision record should state the pause and the evidence required to resume.

Report the distribution of attempts, not only the percentage within the limit. A campaign can have most people at one attempt and a small group at ten. The small group is where the complaints and the risk concentrate. Show the maximum, the median, and the count of people above the rule.

The operational use is to change the reset event, the campaign schedule, or the list quality. If a person is called repeatedly because three systems each think they own the record, the repair is a single customer view, not a lower attempt cap. Name the owner and retest the people-level count.

Sources checked October 22, 2026: FCC, Telemarketing and the Telephone Consumer Protection Act | https://www.fcc.gov/general/telemarketing-and-telephone-consumer-protection-act | National Do Not Call Registry, Business Guidance | https://telemarketing.donotcall.gov/ | NIST Privacy Framework | https://www.nist.gov/privacy-framework. These sources provide regulatory, privacy, and data-reliability context. They do not certify this protocol or any provider's performance, and they do not replace advice from qualified counsel.